Patterning of an extracellular matrix (ECM) layer on silicon by atmosphericpressure plasmas for living neuron network chip fabrication
Ayumi Ando1, Toshifumi Asano2, Hidetaka Uno3, Ryugo Tero4,
Katsuhisa Kitano1, Tsuneo Urisu5 and Satoshi Hamaguchi1
1
Center for Atomic and Molecular Technologies, Graduate School of Engineering, Osaka University, 2-1
Yamadaoka, Suita, Osaka, 565-0871, Japan
2
Division of Neuroscience, Department of Developmental Biology and Neurosciences, Graduate School of Life
Science, Tohoku University, 2-1-1 Katahira, Aoba, Sendai, Miyagi, 980-8577, Japan.
3
4
Department of Structural Molecular Science, The Graduate University for Advanced Studies, 38 Nishigo-Naka,
Myodaiji, Okazaki 444-8585, Japan.
Electronics-Inspired Interdisciplinary Research Institute, Toyohashi University of Technology, 1-1 Hibarigaoka,
Tempakucho, Toyohashi, Aichi, 441-8580, Japan.
5
Division of Biomolecular Sensing, Department of Life and Coordination-Complex Molecular Science, Institute
for Molecular Science, 38 Nishigo-Naka, Myodaiji, Okazaki 444-8585, Japan.
Abstract: A new extracellular matrix (ECM) patterning technique has been developed with the use of lowfrequency (LF) atmospheric pressure plasma jets. Since ECM deposited on a silicon substrate surface can serve as
a scaffold for cells, a micro-patterned ECM layer may be used to arrange living cells in a desired pattern on a
substrate surface. In this study, ECM layers deposited on a silicon substrate were patterned in 100~200m sizes
by LF plasma jet application through a metal stencil mask. When cells were grown on the patterned ECM on
silicon, they aligned along the ECM patterns. The results indicate that the process based on an atmospheric
pressure plasma with a metal stencil mask is a feasible method to pattern a large area of an ECM layer on a silicon
substrate.
Keywords: atmospheric pressure plasma jets, extracellular matrix, patterning, bio chips,
1. Introduction
Technologies for semiconductor manufacturing
can be used to fabricate part of a biochip, which
serves to transmit and receive electrical signals
among cells or biological materials. However, few
biological materials or living cells can be
immobilized directly on a silicon surface, so, in
practice, living cells are typically immobilized on an
extracellular matrix (ECM) layer deposited on a
silicon substrate [1]. The main functions of ECMs
are to serve as a scaffold for cells and to regulate
intercellular communication. The ECM is typically
composed of some types of proteins such as collagen,
laminin, and fibronectin.
For the fabrication of cell chips for diagnosis of
brain diseases such as Alzheimer’s, the formation of
a large-area neuron network on a silicon surface is
considered to be a crucial step. For most living cells
to be immobilized on a silicon surface, a buffer layer
made of certain proteins must be laid out on the
silicon surface. Especially the axon of a neuron,
which transmits electrical signals, typically grows
and extends only along an ECM layer, so a patterned
ECM layer on a silicon chip may be used for the
formation of a neuron network. So far micropatterning of ECMs has been performed mostly by
micro
contact
printing
based
on
polydimethylsiloxane (PDMS) stamps [2, 3].
However it is generally considered that extension of
micro contact printing techniques to large-scale
patterning of ECMs is not feasible and a new
technique for such patterning has been in demand.
Reliability of patterned ECMs with respect to cell
growth for extended periods is also an issue for
ECM patterning.
The goal of this study is to develop an ECM
pattering technique by atmospheric pressure plasmas
(APPs). APPs may be used for such patterning
because of their ability to produce highly reactive
radicals while maintaining low-gas temperatures.
APPs also have an advantage that they can be used
in air without requiring expensive vacuum
equipment. In this study, it has been demonstrated
that application of a low-temperature low-frequency
(LF) atmospheric-pressure plasma jets to a uniform
ECM layer formed on a silicon substrate through a
patterned stencil is a good candidate for ECM
patterning techniques.
2. Experimental set up
LF plasma jets were applied to a 1515 mm2
silicon substrate coated with ECM and a stainlesssteel stencil mask was directly placed on the ECM
layer, as shown in Fig. 1. A SiO2 layer of 1 m in
thickness was formed on the silicon substrate surface
by thermal oxidation (900 ℃, 10 hours, O2+H2O
95 ℃ bubbling) before ECM coating. LF plasma
jets were discharged in air by pulsed high voltages
applied to an electrode that surrounds a glass tube
inside which a helium (He) gas flows. Details of the
system may be found in Ref. [4, 5]. In our
experiments, the pulse frequency of power supply
(1)
(2)
He gas
Glass tube
LF high voltage
supply
Electrode
Stencil mask
Plasma jet
ECM
Silicon substrate
Figure 1. Experimental set up for ECM patterning by LF
plasma jets.
was 10 kHz, the inner diameter of the glass tube was
4 mm, and the distance between the sample substrate
and the end of glass tube was 50 mm. The peak-topeak voltage applied to the electrode (which was
wound around the glass tube), the flow rate of He
gas, and plasma irradiation time were varied in the
process of searching optimal conditions for ECM
patterning.
Two different types of stencil masks were used in
our experiments; one had three different sets of slits
in area of 1 cm2, as shown in Fig. 2; (1) ten 100 mwide slits with a slit interval (i.e., the distance
between two successive slits) of 100 m, (2) ten 100
m-wide slits with a slit interval of 200 m and (3)
ten 200 m-wide slits with a slit interval of 100 m,
while the other had holes about 100 m in diameter
(3)
100m
Figure 2. A stainless-steel mask used in this study. The mask
was pinned down to the substrate by glue.
Figure 3. A micrograph of mask which has holes of about
100 m in diameter with an interval of about 100 m in area
of 1 cm2.
100m
Figure 4. PLL patterns in area (3) of Fig.2 observed by
fluorescence microscope. PLL remaining on the silicon
substrate is shown in white. The dashed white lines indicate the
edge positions of the original mask slits.
100m
Figure 5. PLL patterned with the mask of Fig.3 observed by
fluorescence microscope. PLL remaining on silicon substrate is
shown in white.
3. Experimental results
gas flow rate of 4.5 L/min. It has been observed that,
if LF plasma jets were applied under right conditions
for each mask, PLL deposited on the substrate was
patterned more or less uniformly following the mask
patterns.
(1) Fluorescently-labeled Poly-L-Lysine patterning
by LF plasma jets
(2) Cell culture on a patterned ECM
Silicon substrate which had been spin coated with
a fluorescently-labeled (fluorescein isothiocyanate
labeled) Poly-L-Lysine (FITC-PLL, SIGMA P3545,
500 g/ml) layer was patterned by LF plasma jets.
PLL is small natural homopolymer of the essential
amino acid L-lysine [6]. The molecular weight of
FITC-PLL used in this experiment was 15,00030,000 and the extent of labeling was 0.003-0.01
mol FITC per mol lysine monomer. The substrate
surfaces after patterning were observed by
fluorescence microscopy. Figure 4 shows a
fluorescence micrograph of a silicon surface located
in the area (3) of Fig. 2 when LF plasma jets were
applied for 15 seconds uniformly to the substrate
with a peak-to-peak voltage of 7 kV and a gas flow
rate of 4 L/min [7]. PLL remaining on the substrate
is indicated in white in this micrograph. The dashed
white lines of Fig. 4 delineate the original slit
patterns of the stencil mask. Figure 5 shows a
fluorescence micrograph of a silicon surface
patterned with a mask of holes after LF plasma jets
were applied for 20 seconds uniformly to the
substrate with a peak-to-peak voltage of 7 kV and a
HEK293 cells (Human Embryo Kidney cells)
were cultured on a fibronectin (SIGMA F2006, 200
g/ml) layer patterned by LF plasma jets.
Fibronectin is another type of ECM. Cellular
adherence properties to ECM after patterning were
examined. Types of ECM are selected based on cell
types as certain cells grow only on certain ECMs.
Fibronectin is a multifunctional glycoprotein
composed of subunits with a molecular weight of
approximately 250 kilodaltons (kDa) [8]. The
micrograph of area (3) in Fig. 2 taken 4 days after
the initiation of cell culture is shown in Fig. 6.
Figure 6 shows that cells adhered to and proliferated
along the patterned fibronectin strips although the
observed cell patterns were narrower than the width
of the mask slit. Similar cell adhesion was also
observed in the areas (1) and (2) of Fig. 2 [7].
Similar experiments were also performed for PC12
cells (Rat adrenal pheochromocytoma cell line) on
PLL layer, which are model cells of neuron
differentiation. It was observed that PC12 cells also
adhered to and proliferated along the patterned PLL
strips (data not shown).
with an interval of about 100 m in area of 1 cm2, as
shown in Fig. 3.
method to pattern a large area of ECM layer at least
up to several square centimetres. Clarification of
APP
patterning
mechanisms
(especially
identification of influencing radicals and chemical
reactions) will be deferred to future studies.
Acknowledgements
100m
Figure 6. A micrograph of cell arrangement patterns in the
area (3) of Fig. 2 on the silicon substrate. The dashed white
lines indicate the original mask slit patterns.
(3) Observation of the substrate surface by Fourier
transform infrared spectroscopy (FT-IR) and atomic
force microscopy (AFM)
We also observed the fibronectin surface with
Fourier transform infrared spectroscopy (FT-IR) and
atomic force microscopy (AFM) after being
patterned by LF plasma jets [9].
Fibronectin was deposited on a gold surface of a
substrate. The surface was formed by coating
chromium (a few nanometres thick) and gold (45 nm
thick) on a glass substrate [10]. In FT-IR, the amid I
band spectra (1600 ~ 1700 cm-1) of fibronectin on
the substrate was observed as a function of plasma
application time. For AFM observation, fibronectin
was deposited on a silicon substrate and the surface
profile of fibronectin layer was observed before and
after plasma application for 120 sec. The results of
these observations indicated that chemically reactive
gas-phase species such as free radicals generated by
the plasma were likely to have reacted with the ECM
and removed it from the surface. The reactive
species, however, have not been identified. The
analysis of patterning mechanisms by low
temperature APPs will be deferred to future studies.
4. Conclusions
LF plasma jets have been used to pattern ECM
layers deposited on Si substrates in 100 ~ 200 m
sizes. It has been observed that human cells and
model cells of neuron differentiation can adhere to
and proliferate on the patterned fibronectin layers.
These results indicate that a process based on APP
application with a metal stencil mask is a feasible
This research has been funded in part by the
Grants-in-Aid for Scientific Research from the
Ministry of education, culture, sports, science and
technology (MEXT) of Japan and the Joint Studies
Program (2010 ~ 2011) of the Institute for Molecular
Science.
References
[1] T. Urisu, T. Asano, Z. Zhang, H. Uno, R. Tero,
H. Junkyu, H. Isoda, Y. Arima, H. Iwata, K.
Shibasaki and M. Tominaga: Anal. Bioanal. Chem.
391 (2008) 2703.
[2] A. Bernard, E. Delamarche, H. Schmid, B.
Michel, H. R. Bosshard and H. Biebuyck: Langmuir
14 (1998) 2225.
[3] M. Nishizawa, H. Kaji, K. Takoh and T. Matsue:
J. Surf. Sci. Soc. Jpn. 25 (2004) 290.
[4] Kitano K and Hamaguchi S 2008 The Japan
Society of Applied Physics 77(4) 383
[5] S. Ikawa, K. Kitano, and S. Hamaguchi: Plasma
Process. Polym. 7 (2010) 33.
[6] S. Shima, and H. Sakai, Agric . Biol. Chem. 41
(1977) 1807.
[7] A. Ando, T. Asano, R. Tero, K. Kitano, T. Urisu
and S. Hamaguchi: Proceedings of 7th International
Conference on Reactive Plasmas, 28th Symposium
on Plasma Processing and 63rd Gaseous
Electronics Conference (ICRP-7/SPP-28/GEC-63),
KWP.034, 2010.
[8] S K Akiyama, K Nagata and K M Yamada,
Biochim. Biophys. Acta 1031 (1990) 91.
[9] A. Ando, M. A. Sayed, T. Asano, R. Tero, K.
Kitano, T. Urisu and S. Hamaguchi: J. Phys. Conf.
Series 232 (2010) 012019.
[10] M. A. Sayed, H. Uno, K Harada, K. Tanaka, H.
Y. Kim, Y. Nakaoki, K. Okumura, R. Tero and T.
Urisu, Chem. Phys. Lett. 46 (2008) 235.